Iranian and Hungarian EFL Students’ Essays. A Corpus-Based Study

Abstract

The study of learner language, mostly written, has become a major field within corpus linguistics in the past twenty years. Since the beginnings in the early 1990, with the development of the ICLE and the JPU Corpus, the field of learner corpus research (LCR) has established a number of standards specific to the design, analysis, and application of such corpora (Granger, 1998; Horváth, 2001, 2015). This dissertation aims to present the results of what can be considered a new vista in LCR: the qualitative and quantitative investigation of diachronic features of written language. I was interested in whether and how learner language has changed since the early years of the BA program, introduced in Hungary in 2006, on the one hand, and how the learner language in the BA programs in Hungary and Iran differ from each other. I have developed two small corpora; the first one is the Happy Corpus consisting of essays written at the University of Pécs as part of English majors’ proficiency exams, and the second one is the HI Corpus encompassing the essays written by the Hungarian students at the University of Pécs and the Iranian students at Amol Islamic Azad University, who attended English writing courses. The Happy Corpus has two components: the 2009 and the 2014 batches. In both exam situations, students were presented with the same choice of two themes, of which they were instructed to develop one essay, of about 300 words. I hypothesized that of the two topics, “Don’t Worry; Be Happy” would be chosen by an overwhelming majority of the students in 2014, as this had been the case in 2009, too. Each of the Hungrian and Iranian subcorpora in the HI Corpus has five components: the Anxiety, the Fantasy, the Memories, the Favorite English Texts, and the Purpose of Studyin English. I investigated content and language features of the subcorpora in the Happy Corpus and the HI Corpus separately. Specifically, the study aimed to examine whether there were distinguishing differences between the 2009 and 2014 subcorpora in the Happy Corpus, and the Hungarian an Iranian subcorpora in the HI Corpus thematically and lexically, two aspects that correspond to assessment criteria applied in the exam. In this dissertation, I will highlight the most interesting results and suggest ways in which the results can inform future development of essay writing skills as well as testing procedures.